Linanthus parryae, a diminutive desert annual with white or blue flowers, has been the focus of a long-standing debate among evolutionary biologists. At issue is whether the flower color polymorphism in this species is the product of random genetic drift, as Sewall Wright argued, or of natural selection, as proposed by Carl Epling and his colleagues. Our long-term studies of three polymorphic populations in the Mojave Desert demonstrate that flower color is subject to selection that varies in both time and space in its direction and magnitude. For all sites taken together, blue-flowered plants produced more seeds than white-flowered plants in years of relatively low seed production, whereas white-flowered plants had higher fitness in years of high seed production. Evidence of selection on flower color was found in two of the three study sites. Differences in fitness between the color morphs were sometimes large, with selection coefficients as high as 0.60 in some years. Our longest period of observations was at Pearblossom site 1, where plants reached appreciable densities in seven of the 11 years of study. Here we found significant differences in the seed production of the color morphs in six years, with three years of blue advantage and three years of white advantage. For all sites taken together, total spring precipitation (March and April) was positively correlated with both absolute and relative seed production of the color morphs. At Pearblossom site 1, blue-flowered plants typically had a fitness advantage in years of low spring precipitation, whereas white-flowered plants had a fitness advantage in years of high spring precipitation. This temporal variation in selection may contribute to the maintenance of the flower-color polymorphism at Pearblossom site 1, whereas gene flow from neighboring populations is proposed as the principal factor maintaining the polymorphism at the other study sites. We found no significant differences between the color morphs in pollinator visitation rate or in their carbon isotope ratio, a measure of water-use efficiency. Although the mechanism of selection remains elusive, our results refute Wright's conclusion that the flower color polymorphism in L. parryae is an example of isolation by distance, a key component of his shifting balance theory of evolution.

Corresponding Editor: C. Eckert

Motivated by data demonstrating fluctuating relative and absolute fitnesses for white- versus blue-flowered morphs of the desert annual Linanthus parryae, we present conditions under which temporally fluctuating selection and fluctuating contributions to a persistent seed bank will maintain a stable single-locus polymorphism. In L. parryae, blue flower color is determined by a single dominant allele. To disentangle the underlying diversity-maintaining mechanism from the mathematical complications associated with departures from Hardy-Weinberg genotype frequencies and dominance, we successively analyze a haploid model, a diploid model with three distinguishable genotypes, and a diploid model with complete dominance. For each model, we present conditions for the maintenance of a stable polymorphism, then use a diffusion approximation to describe the long-term fluctuations associated with these polymorphisms. Our protected polymorphism analyses show that a genotype whose arithmetic and geometric mean relative fitnesses are both less than one can persist if its relative fitness exceeds one in years that produce the most offspring. This condition is met by data from a population of L. parryae whose white morph has higher fitness (seed set) only in years of relatively heavy rain fall. The data suggest that the observed polymorphism may be explained by fluctuating selection. However, the yearly variation in flower color frequencies cannot be fully explained by our simple models, which ignore age structure and possible selection in the seed bank. We address two additional questions—one mathematical, the other biological—concerning the applicability of diffusion approximations to intense selection and the applicability of long-term predictions to datasets spanning decades for populations with long-lived seed banks.

Corresponding Editor: O. Savolainen

We incorporate a representation of Plasmodium falciparum recombination within a discrete-event model of malaria transmission. We simulate the introduction of a new parasite genotype into a human population in which another genotype has reached equilibrium prevalence and compare the emergence and persistence of the novel recombinant forms under differing cross-reactivity relationships between the genotypes. Cross-reactivity between the parental (initial and introduced) genotypes reduces the frequency of appearance of recombinants within three years of introduction from 100% to 14%, and delays their appearance by more than a year, on average. Cross-reactivity between parental and recombinant genotypes reduces the frequency of appearance to 36% and increases the probability of recombinant extinction following appearance from 0% to 83%. When a recombinant is cross-reactive with its parental types, its probability of extinction is influenced by cross-reactivity between the parental types in the opposite manner; that is, its probability of extinction after appearance decreases. Frequencies of P. falciparum outcrossing are mediated by frequencies of mixed-genotype infections in the host population, which are in turn mediated by the structure of cross-reactivity between parasite genotypes. The three leading hypotheses about how meiosis relates to oocyst production lead to quantitative, but no qualitative, differences in these results.

Corresponding Editor: M. Riley

The existence of parasitic constraints on the evolution of life-history traits in free-living organisms has been demonstrated in several plant and animal species. However, the association between different diseases and human traits is virtually unknown. We conducted a comparative analysis on a global scale to test whether the diversity of human diseases, some of them responsible for high incidences of morbidity and mortality, were associated with host life-history characteristics. After controlling for direct confounding effects exerted by historical, spatial, economic, and population patterns and their interactions, our findings show that human fertility increases with the diversity and structure of disease types. Thus, disease control may not only lower the costs associated with morbidity, but could also contribute directly or indirectly to reductions in human population growth.

Corresponding Editor: D. Waller

Rhexia, with 11 species in the Coastal Plain province of North America, is the only temperate zone endemic of the tropical eudicot family Melastomataceae. It is a member of the only pantropical tribe of that family, Melastomeae. Based on the chloroplast gene ndhF, we use a fossil-calibrated molecular clock to address the question of the geographic origin and age of Rhexia. Sequences from 37 species in 21 genera representing the tribe's geographical range were analyzed together with five outgroups. To obtain better clade support, another chloroplast region, the rpl16 intron, was added for 24 of the species. Parsimony analysis of the combined data and maximum-likelihood analysis of ndhF alone indicate that the deepest split is between Rhexia plus its sister group, a small Central American genus, and all other Melastomeae. Old World Melastomeae are monophyletic and nested within New World Melastomeae. Although likelihood-ratio tests of clock and nonclock substitution models for the full or moderately pruned datasets rejected the clock, these models yielded identical topologies (for 30 taxa) with few significantly different branch lengths as assessed by a Student's t-test. Age estimates obtained were 22 million years ago (Mya) for the divergence of Rhexia from its sister group, 12 Mya for the dispersal of Melastomeae from the New World to West Africa, and 1 Mya for the diversification of Melastoma in Southeast Asia. The only other genus of Melastomeae to have reached Southeast Asia from Africa or Madagascar is Osbeckia. The age and geographic distribution of fossils, which come from Miocene sites throughout Eurasia, suggest that Melastomeae once ranged from Eurasia across Beringia to North America from whence they reached South America and subsequently Africa and Southeast Asia. Climate deterioration led to their extinction in the Northern Hemisphere, with Rhexia possibly surviving in Coastal Plain refugia.

Corresponding Editor: D. Baum

Hybrids have long been recognized as a potential pathway for gene flow between species that can have important consequences for evolution and conservation biology. However, few studies have demonstrated that genes from one species can introgress or invade another species over a broad geographic area. Using 35 genetically mapped restriction fragment length polymorphism (RFLP) markers of two species of cottonwoods (Populus fremontii x P. angustifolia) and their hybrids (n = 550 trees), we showed that the majority of the genome is prohibited from introgressing from one species into the other. However, this barrier was not absolute; Fremont cpDNA and mtDNA were found throughout the geographic range of narrowleaf cottonwood, and 20% of the nuclear markers of Fremont cottonwood introgressed varying distances (some over 100 km) into the recipient species' range. Rates of nuclear introgression were variable, but two nuclear markers introgressed as fast as the haploid, cytoplasmically inherited chloroplast and mitochondrial markers. Our genome-wide analysis provides evidence for positive, negative, and neutral effects of introgression. For example, we predict that DNA fragments that introgress through several generations of backcrossing will be small, because small fragments are less likely to contain deleterious genes. These results argue that recombination will be important, that introgression can be very selective, and that evolutionary forces within the hybrid population to effectively “filter” gene flow between species. A strong filter may make introgression adaptive, prevent genetic assimilation, lead to relaxed isolating mechanisms, and contribute to the stability of hybrid zones. Thus, rather than hybridization being a negative factor as is commonly argued, natural hybridization between native species may provide important genetic variation that impacts both ecological and evolutionary processes. Finally, we propose two hypotheses that contrast the likelihood of contemporary versus ancient introgression in this system.

Corresponding Editor: S. Tonsor

In the northern acorn barnacle, Semibalanus balanoides, genotype frequencies of three genetic markers were tracked over time in four types of intertidal habitats. These habitats were selected to represent natural variation in several environmental parameters, specifically the degree of physical stress experienced by barnacles. Frequencies for one allozyme locus (Gpi) and a presumably neutral mtDNA marker were homogeneous among habitats in each temporal sample. Similarly, no temporal stratification in genotype frequencies was evident across the five sampling intervals: from planktonic larvae sampled in March to juveniles collected at the end of June. In contrast to the Gpi and mtDNA loci, Mpi genotypes significantly changed in frequency in two habitats in the high intertidal zone. On exposed substrate, the Mpi-FF homozygote increased in frequency, whereas the alternative homozygote, Mpi-SS, significantly decreased in frequency. Barnacles that were protected from environmental stress at high intertidal heights by the Ascophyllum nodosum algal canopy demonstrated the opposite pattern. In both habitats, the change in frequency of the heterozygote was intermediate to that of the homozygous genotypes. Furthermore, these patterns of genotype-by-environment association reflected a pulse of genotype-specific mortality that occurred over a two-week interval subsequent to metamorphosis from the larval to the adult form. These data indicate that each Mpi homozygote is the highest fitness genotype in some portion of the intertidal environment. Using the Levene (1953) model to evaluate the spatial variation in genotypic fitness, the stable maintenance of the Mpi polymorphism is predicted under certain subsets of conditions. Environmental heterogeneity in the intertidal zone translates to spatial variation in selection pressures, which may result in the active maintenance of the Mpi polymorphism in this species.

Corresponding Editor: R. Burton

Studies of patterns of molecular variation in natural populations can provide important insights into a number of evolutionary problems. Among these, the question of whether geographic factors are more important than ecological factors in promoting population differentiation and ultimately speciation has been an important and contentious area in evolutionary biology. Systems involving herbivorous insects have played a leading role in this discussion. This study examined the distribution of molecular variation in a highly specialized gall-forming insect, grape phylloxera (Daktulosphaira vitifoliae Fitch), that is found on both sympatric and allopatric host-plant species of the genus Vitis. In addition, the relationship of insects in the introduced range in the United States to ancestral populations in the native range was examined. Evidence for differentiation along host-plant lines from both nuclear (RAPD) and mitochondrial (COI) DNA was confounded with the effect of geography. Differentiation was found where hosts were allopatric or parapatric, but no evidence was found for such differentiation on two hosts, V. vulpina and V. aestivalis, that are broadly sympatric. The question of population differentiation onto these sympatric hosts can be considered to be resolved—it has not occurred in spite of a long history of association. Evidence was equivocal, but suggestive of a period of divergence in allopatry prior to reestablishment of contact, for insects associated with another host plant species, V. cinerea, found in both sympatric and parapatric populations. A low level of diversity and placement of samples collected from the grape species V. riparia at the tip of a phylogenetic tree supports the hypothesis that this host has been recently colonized from populations from the Mississippi Valley. A polyphyletic origin for biotype B grape phylloxera was supported: Although most samples collected from vineyards in the introduced range in California had similar haplotypes, they were closely related to natives on V. vulpina from the Atlantic Coast–Piedmont region. All samples collected from vineyards in Oregon and Washington were closely related to natives on V. riparia in the northern United States.

Corresponding Editor: C. Boggs

Studies on selection for faster development in Drosophila have typically focused on the trade-offs among development time, adult weight, and adult life span. Relatively less attention has been paid to the evolution of preadult life stages and behaviors in response to such selection. We have earlier reported that four laboratory populations of D. melanogaster selected for faster development and early reproduction, relative to control populations, showed considerably reduced preadult development time and survivorship, dry weight at eclosion, and larval growth rates. Here we study the larval phase of these populations in greater detail. We show here that the reduction in development time after about 50 generations of selection is due to reduced duration of the first and third larval instars and the pupal stage, whereas the duration of the second larval instar has not changed. About 90% of the preadult mortality in the selected populations is due to larval mortality. The third instar larvae, pupae, and freshly eclosed adults of the selected populations weigh significantly less than controls, and this difference appears during the third larval instar. Thereafter, percentage weight loss during the pupal stage does not differ between selected and control populations. The minimum amount of time a larva must feed to subsequently complete development is lower in the selected populations, which also exhibit a syndrome of reduced energy expenditure through reduction in larval feeding rate, larval digging and foraging activity, and pupation height. Comparison of these results with those observed earlier in populations selected for adaptation to larval crowding and faster development under a different protocol from ours reveal differences in the evolved traits that suggest that the responses to selection for faster development are greatly affected by the larval density at which selection acts and on details of the selection pressures acting on the timing of reproduction.

Corresponding Editor: C. López-Fanjul

Eye stalks and their scaling relationship with body size are important features in the mating system of many diopsid species, and sexual selection is a critical force influencing the evolution of this exaggerated morphology. Interspecific variation in eye span suggests there has been significant evolutionary change in this trait, but a robust phylogenetic hypothesis is required to determine its rate and direction of change. In this study, the pattern of morphological evolution of eye span is assessed in a phylogenetic framework with respect to its function in the sexual system of these flies. Specifically, we examine within the family Diopsidae the pattern of increase and decrease in sexual dimorphism, the morphological coevolution of eye span between males and females, and the evolutionary flexibility of eye-span allometry. Based on several different methods for reconstructing morphological change, results suggest a general pattern of evolutionary flexibility, particularly for eye-span allometry. Sexual dimorphism in eye span has evolved independently at least four times in the family and this trait also has undergone several reductions within the genus Diasemopsis. Despite most species being dimorphic, there is a strong phylogenetic correlation between males and females for mean eye span. The coevolution between the sexes for eye-span allometry, however, is significantly weaker. Overall, eye-span allometry exhibits significantly more change on the phylogeny than the other morphological traits. The evolutionary pattern in eye-span allometry is caused primarily by changes in eye-span variance. Therefore, this pattern is consistent with recent models that predict a strong relationship between sexual selection and the variance of ornamental traits and highlights the significance of eye-span allometry in intersexual and intrasexual signaling.

Corresponding Editor: K. Ross

Male crickets (Teleogryllus oceanicus) produce a complex call consisting of two elements, the long chirp (three to eight sound pulses) followed by a series of short chirps (each with two sound pulses). There is significant geographic variation in the temporal structure of calls, and the long chirp is selected against by acoustically orienting parasitoids in some populations. Here we examine geographic variation in female preference functions for the amount of long chirp. In general, females prefer calls with greater proportions of long chirp, although the strength and nature of selection varied across populations. Variation in preference functions did not match variation in call structure. There was a mismatch between the proportion of long chirp produced by males in a population and the proportion of long chirp preferred by females. The convergent preferences of predators and females are likely to maintain genetic variation in song traits in parasitized populations. The apparent mismatch between preference and trait is discussed in relation to theoretical models of preference evolution.

Corresponding Editor: D. Wheeler

New Zealand's isolation, its well-studied rapidly changing landscape, and its many examples of rampant speciation make it an excellent location for studying the process of genetic differentiation. Using 1520 base pairs of mitochondrial DNA from the cytochrome oxidase subunit I, ATPase subunits 6 and 8 and tRNA^Asp genes, we detected two well-differentiated, parapatrically distributed clades within the widespread New Zealand cicada species Maoricicada campbelli that may prove to represent two species. The situation that we uncovered is unusual in that an ancient lineage with low genetic diversity is surrounded on three sides by two recently diverged lineages. Using a relaxed molecular clock model coupled with Bayesian statistics, we dated the earliest divergence within M. campbelli at 2.3 ± 0.55 million years. Our data suggest that geological and climatological events of the late Pliocene divided a once-widespread species into northern and southern components and that near the middle of the Pleistocene the northern lineage began moving south eventually reaching the southern clade. The southern clade seems to have moved northward to only a limited extent. We discovered five potential zones of secondary contact through mountain passes that will be examined in future work. We predict that, as in North American periodical cicadas, contact between these highly differentiated lineages will exist but will not involve gene flow.

Corresponding Editor: D. Waller

Morphological differentiation in the ground beetles of the Nebria gregaria group, found on the Queen Charlotte Islands, has been used as support for the glacial refugium proposed for the northwest coast of North America. Two members of this species group, N. charlottae and N. louiseae, are restricted to cobble beaches in this archipelago. A third, N. haida, is found only in alpine regions of the archipelago and the adjacent mainland. The remaining two species of the gregaria group, N. lituyae and N. gregaria, show highly restricted distributions in the mountains of the Alaska panhandle and on the beaches of the Aleutian Islands, respectively. To determine the relationships of the five species, we conducted phylogenetic analyses on nucleotide sequence data obtained from five regions of the mitochondrial DNA. In total, 1835 bp were analyzed. The results suggest that one species, N. lituyae, does not belong in the gregaria group, and that only seven mutations separated the two most divergent of the four remaining species. We also conducted random amplified polymorphic DNA fingerprinting analyses on genomic DNA extracted from the five species. Analyses of genetic diversity revealed a lack of molecular differentiation among the Queen Charlotte species, suggesting that these populations may be postglacial in origin and that together N. gregaria, N. charlottae, N. louiseae, and N. haida might represent local variations of a single species. These results are consistent with conclusions derived for the morphological and genetical differentiation among Gasterosteus populations in the archipelago.

Corresponding Editor: S. Karl

The prevalence and meaning of morphological trends in the fossil record have undergone renewed scrutiny in recent years. Studies have typically focused on trends in body size evolution, which have yielded conflicting results, and have only rarely addressed the question as to whether other morphological characteristics show persistent directionality over long time scales. I investigated reduction in number of skull and lower jaw bones (through loss or fusion) over approximately 150 million years of premammalian synapsid history. The results of a new skull simplification metric (SSM), which is defined as a function of the number of distinct elements, show that pronounced simplification is evident on both temporal (i.e., stratigraphic) and phylogenetic scales. Postcranial evolution exhibits a similar pattern. Skull size, in contrast, bears little relationship with the number of distinct skull bones present.

Synapsid skulls carried close to their observed maximum number of elements for most of the Late Carboniferous and Early Permian. The SSM decreased in the Late Permian but, coincident with the radiation of early therapsids, the range of observed SSM values widened during this interval. From derived nonmammalian cynodonts in the Early Triassic through the earliest mammals in the Early Jurassic, both the minimum and maximum SSM decreased. Data from three representative modern mammals (platypus, opossum, and human) suggest that this trend continues through the Cenozoic.

In a phylogenetic context, the number of skull elements present in a taxon shows a significant negative relationship with the number of branching events passed from the root of the tree; more deeply embedded taxa have smaller SSM scores. This relationship holds for various synapsid subgroups as well. Although commonly ascribed to the effects of long-term selection, evolutionary trends can alternatively reflect an underlying intrinsic bias in morphological change. In the case of synapsid skull bones (and those of some other tetrapods lineages), the rare production of novel, or neomorphic, elements may have contributed to the observed trend toward skeletal simplification.

Corresponding Editor: M. Zelditch

In western Panama, an unusual hybrid zone exists between white-collared manakins, Manacus candei, and golden-collared manakins, M. vitellinus. Unidirectional introgression of plumage traits from vitellinus into candei has created a region in which all definitively plumaged males have a collar that is lemon-colored. These males are nearly indistinguishable from white-collared candei genetically and morphometrically, but strongly resemble golden-collared vitellinus due to the introgression of secondary sexual plumage traits, particularly the lemon-colored collar. The introgression could be explained by sexual selection for golden-collared traits or by a series of mechanisms that do not invoke sexual selection (e.g., neutral diffusion, dominant allele). Sexual selection on male-male interactions implies behavioral differences among the plumage forms—specifically that golden- and lemon-collared males should be more aggressive than white-collared males. In contrast, the nonsexual hypotheses predict behavioral similarity between lemon- and white-collared males, based on their nearly identical genetics. We tested the sexual selection hypothesis experimentally, by presenting males with taxidermic mounts of the three forms. As response variables, we monitored vocalizations and attacks on the mounts by replicate subject males. Both golden-collared and lemon-collared males were more likely to attack than were white-collared males, as predicted under sexual selection but not by the nonsexual hypotheses. Lemon-collared males were more vocally reactive than either parental form, contrary to the prediction of the nonsexual hypotheses. Our study demonstrates that sexual selection on male-male interactions may play an important role in the dynamics of character evolution and hybrid zones.

Corresponding Editor: B. Crespi

Although sexual selection through female choice explains exaggerated male ornaments in many species, the evolution of the multicomponent nature of most sexual displays remains poorly understood. Theoretical models suggest that handicap signaling should converge on a single most informative quality indicator, whereas additional signals are more likely to be arbitrary Fisherian traits, amplifiers, or exploitations of receiver psychology. Male nuptial plumage in the highly polygynous red-collared widowbird (Euplectes ardens) comprises two of the commonly advocated quality advertisements (handicaps) in birds: a long graduated tail and red carotenoid coloration. Here we use multivariate selection analysis to investigate female choice in relation to male tail length, color (reflectance) of the collar, other aspects of morphology, ectoparasite load, display rate, and territory quality. The order and total number of active nests obtained are used as measures of male reproductive success. We demonstrate a strong female preference and net sexual selection for long tails, but marginal or no effects of color, morphology, or territory quality. Tail length explained 47% of male reproductive success, an unusually strong fitness effect of natural ornament variation. Fluctuating tail asymmetry was unrelated to tail length, and had no impact on mating success. For the red collar, there was negative net selection on collar area, presumably via its negative relationship with tail length. None of the color variables (hue, chroma, and brightness) had significant selection differentials, but a partial effect (selection gradient) of chroma might represent a color preference when tail length is controlled for. We suggest that the red collar functions in male agonistic interactions, which has been strongly supported by subsequent work. Thus, female choice targets only one handicap, extreme tail elongation, disregarding or even selecting against the carotenoid display. We discuss whether long tails might be better indicators of genetic quality than carotenoid pigmentation. As regards the evolution of multiple ornaments, we propose that multiple handicap signaling is stable not because of multiple messages but because of multiple receivers, in this case females and males.

Corresponding Editor: P. Dunn

Predictions associated with opposing selection generating minimum variance in basal metabolic rate (BMR) in mammals at a constrained body mass (CBM; 358 g) were tested. The CBM is presumed to be associated with energetic constraints linked to predation and variable resources at intermediate sizes on a logarithmic mass scale. Opposing selection is thought to occur in response to energetic constraints associated with predation and unpredictable resources. As body size approaches and exceeds the CBM, mammals face increasing risks of predation and daily energy requirements. Fast running speeds may require high BMRs, but unpredictable and low resources may select for low BMRs, which also reduce foraging time and distances and thus predation risks. If these two selection forces oppose each other persistently, minimum BMR variance may result. However, extreme BMR outliers at and close to the CBM should be indicative of unbalanced selection and predator avoidance alternatives (escapers vs. defenders), and may therefore provide indirect support for opposing selection. It was confirmed that body armor in defenders evolves at and above the CBM, and armored mammals had significantly lower BMRs than their nonarmored counterparts. However, analyses comparing the BMR of escapers—the fastest nonarmored runners (Lagomorpha)—with similar-sized counterparts were inconclusive and were confounded by limb morphology associated with speed optimization. These analyses suggest that the risks and costs of predation and the speed limitations of the plantigrade foot may constrain the evolution of large body sizes in plantigrade mammals.

Corresponding Editor: T. Garland Jr.

The Accretion model of Neandertal evolution specifies that this group of Late Pleistocene hominids evolved in partial or complete genetic isolation from the rest of humanity through the gradual accumulation of distinctive morphological traits in European populations. As they became more common, these traits also became less variable, according to those workers who developed the model. Its supporters propose that genetic drift caused this evolution, resulting from an initial small European population size and either complete isolation or drastic reduction in gene flow between this deme and contemporary human populations elsewhere. Here, we test an evolutionary model of gene flow between regions against fossil data from the European population of the Middle and Late Pleistocene. The results of the analysis clearly show that the European population was not significantly divergent from its contemporaries, even in a subset of traits chosen to show the maximum differences between Europeans and other populations. The pattern of changes, over time within Europe of the traits in this subset, does not support the Accretion model, either because the characters did not change in the manner specified by the model or because the characters did not change at all. From these data, we can conclude that special phenomena such as near-complete isolation of the European population during the Pleistocene are not required to explain the pattern of evolution in this region.

Corresponding Editor: E. Brodie III

I document a genetic basis for parallel evolution of life-history phenotypes in the livebearing fish Brachyrhaphis rhabdophora from northwestern Costa Rica. In previous work, I showed that populations of B. rhabdophora that co-occur with predators attain maturity at smaller sizes than populations that live in predator-free environments. I also demonstrated that this pattern of phenotypic divergence in life histories was independently repeated in at least five isolated drainages. However, life-history phenotypes measured from wild-caught fish could be attributed to environmental effects rather than to genetic differences among populations. In the present study, I reared male fish from four populations (two that co-occur with predators and two from predator-free environments) under four sets of environmental conditions. The pattern of phenotypic divergence in maturation size documented in the field between populations collected from different predation environments persisted after two generations in the laboratory. I also found a genetic basis for differences between populations in the age at which males attain maturity and in growth rates. By rearing fish in four different common environments, I tested for phenotypic plasticity in male life-history traits in response to nonlethal exposure to predators. There was a significant delay in the onset of sexual maturity in fish exposed to predators relative to those in the control, but no differences among treatments in size at maturity or growth rates. These results, coupled with previous work on B. rhabdophora, demonstrate a repeated pattern of parallel evolutionary divergence among genetically isolated populations that is strongly associated with predation.

Corresponding Editor: J. Losos

Corresponding Editor: D. Futuyma

Corresponding Editor: D. Futuyma